page
1
2
3
4
downloads
Enclosure Mock-up
July 22, 2008
When thinking of an enclosure for the GMS, it occurred to me to call my cousin Wink.
I was imagining a piece of PVC pipe or a square stock of some sort, and Wink is a very practical handyman plus works with pipes as owner of a heating and air conditioning business.
I'm glad I called, Wink shared some great thoughts that will help make the project happen smoothly.
For one thing, Wink suggested making a prototype of the enclosure out of some inexpensive material such as plastic sheet or cardboard.
As you can see in the photo to the right, I decided to try cardboard.
Actually I traced out the shape of the carboard piece on a sheet of printer paper and then traced it onto the cardboard by punching holes along the lines with the pen tip.
It worked OK, and I'm not being super-exact here, it's just a prototype of a prototype, eh?
One thing that occurred to me is that I want to make it so that I can get it on and off by simply removing the two jack plate screws. I don't want to have to unsolder the wires going to the jack plate if I don't have to, "if it ain't broke, don't fix it", right?
So this will require some geometric attention to detail.
Another factor is that it would be best to support the jack plate fully around it's periphery.
If the final appearance has an elevated section of the jack plate's rim, I don't think that would be too professional.
Then there is the matter of aligning the holes which are not accurate.
And more importantly, we must decide on which side of the jack plate to mount the electronics.
It's a bit tight on the top but spacious on the bottom.
I had originally thought to attach the electronics on top - that is, between the jack plate and the tone dials.
But I just picked up the guitar and strapped it on, testing for the best hand position.
That made it clear - the electronics should go on the bottom side, or opposite side of the tone dials.
To represent the circuitry box of the enclosure, I cut a pill bottle in half with a utility knife and taped it together, forming an example half pipe with closed ends.
This was also convenient because the sections could be overlapped, forming a telescoping length adjustment; just the thing for imagineering.
Then I sized things up, placed the half-pipe with tape, and cut away the excess cardboard.
The result was the enclosure mock-up shown in the image to the right.
Obviously the cardboard needs to be larger to get those corners that are missing, but you get the idea.
it's a mock-up, not a final result.
I just got up and tried on the guitar, feeling for the position of the controls.
Surprisingly it works quite well already, it fits under the tremelo bar even allowing for additional switch height, the three or four buttons are easily accessible, and my hand reaches it conveniently.
The angle is slightly off, though, I may rotate the next one so that it is more vertical and not in parallel with the jack plate, if only so that my hand lands right on the controls at the perfect angle.
So there you have it, our first mock-up.
It was worth a little time to slap something like this together just to get a feel for the physical aspects of things.
Enclosure Mock-up 2
Pleased with the crude cardboard mock-up, I wanted to create something a little bit stronger that wouldn't fall apart if I carried it around in my pocket while shopping for parts.
I looked more closely at the U-HID Nano from Ultimarc and concluded that the pill bottle enclosure was larger than necessary, and I found a small block of wood a bit closer to the correct size.
Then I traced out the cardboard cutout onto a sheet of card stock that I found, and drew a pleasing shape around it.
I then cut out the card stock and taped the wood block to it, then taped the whole thing to my guitar.
This produced the photo to the right.
Of course, it still isn't sturdy enough to carry around so I found some metal to cut out.
I'm using a Dremel tool with a diamond blade to cut out the bottom of a cookie tin for the plate, then I'll glue the wooden block to it with some E6000.
The only thing holding that up is that it's getting late and I don't want to bother the neighbors with noise. So I'll wait until tomorrow to finish the second mock-up.
Notice that I created an opening on the thin edge of the base by the jack plate.
This is so that the wires can be slipped through the slot, making the installation require no soldering on the guitar.
In approaching this installation, I feel most comfortable making the fewest possible modifications to the guitar itself, just for lack of familiarity if nothing else.
Also, I'm trying to accommodate non-technical musicians who might not be comfortable with soldering on the guitar itself.
Well, it's a minor detail but it's best to think these details through early if we can.
Also notice that I have changed the angle of enclosure to better align with my hand while operating the configuration and / or freeze buttons.
I'm not quite pleased with this angular change because it will cause the USB cable to emerge from the guitar at an angle that is pointing away from the guitar cable direction, instead of parallel or pointing toward it.
I may consider repositioning the enclosue and mounting the buttons offset or some trick like that to get the best of both worlds, we'll see.
I think the wooden block might be a little bit shorter than required, but that's OK, it's getting closer to the right size.
One step at a time.
Huh, I just realized something while reviewing the above photo.
Sometimes I like to just kind of stare at things and think about them and ideas occur.
So anyway what I realized was that the GMS base plate can attach to the *top* of the jack plate, not underneath it.
This solves any issues with irregular guitar surfaces (curved profiles), marring the finish, and such concerns because the GMS will be elevated slightly above the surface of the guitar.
It also simplifies installation in that it is no longer necessary to pull out the jack plate.
The base plate will have to be shaped in some way, perhaps countersunk, that accepts the wood screw properly.
So tomorrow I'll revise the shape and cut it out of thin sheet metal, keep you posted.
OK, it's 1am and I found a pair of tin snips to cut the bottom plate, forgot I had them.
Cutting thin sheet metal with a dremel tool was tricky business anyway, the snips are better for this.
I used the snips to crudely munch at the shape, then drilled the screw holes and smoothed the edges with a Dremel and some sandpaper.
After mucking with it like that for a while, I finally came up with the morphy crescent moon shape shown in the photo.
Then with a little E6000 I glued the wooden block into place.
I looked around for some suitable buttons and found these old laser pointer heads to represent buttons.
They give the project a kind of a retro 1940's switch panel look, or something like that.
Finally, the block of wood is actually cut into a U-shape, so there is a place to put a sample USB cable, what luck.
So we've evolved the design a little in a few hours of relaxed playing around with it, and it's good enough to illustrate the product to folks.
Actually, now that I've morphed it this far I'll morph one step more.
Let's say we stuck a powerful magnet from Gauss Boys' super magnet store and put it in the end of the electronics enclosure.
That thing would go "clink!", snapping onto the flat surface of the jack plate above the cable.
With that in mind, we could make a cable assembly that attaches to the guitar cable for radial alignment, then snaps itself onto the guitar with a magnet.
I think you get the idea, so the cable pair and the electronics are all morphed into one magnetically attached unit.
Kind cool, eh?
That solution would involve having the sensor board mounted to the cable, thus removing the need to fasten two connectors all the time, and also not modifying the guitar at all.
Hmmm, this way the product travels with the cable, not the guitar.
Obviously I'm thinking out loud here, just throwing ideas into the mix.
The magnetic cable dealie is another way to go.
Betty Working
You may have noticed in that GMS 10 video that Betty was sort of working but not quite.
Well, I struggled with it for most of a day and finally fixed the problem.
Now Betty's voice is clear as a bell and provides you with feedback unless you drown her out with volume.
I currently have Betty hooked up to the dac so that her voice is unamplified and undistorted.
It might be better to hook betty up to the input of the effects chain so that any gain applied to the audio amplifies her voice as well.
Plus it would be a good indicator of how the sound will pass thru the effects.
Well, we will see if I make that change or not.
When it comes to little things like this, if you have an opinion about it please feel free to send me an email about it.
I try to be responsive to user suggestions where possible.
So anyway, you have a working directory that is specified in the preferences of miniAudicle, be sure you are aware of it and it is pointed to the same directory in which you keep your ChucK programs.
Then create a subdirectory called "Betty" and put all the wav files in "Betty".
That's all you need to do, if you don't hear Betty's voice then make sure the directory is right, that you have at least GMS11.ck, and that the miniAudicle is pointed to the ChucK directory.
Oh, yeah, by the way on the downloads page I have Released GMS11.ck, which you need to run Betty.
So be sure to get that too, they are under the same section heading at the bottom of the page.
I'd rather add more between each release, but it's best that you have the latest version, if only so you can help steer the project if you like.
Right now I'm waiting on a few things, like the imminent release of the next revision of miniAudicle.
This is critical because it may or may not have cross-platform MAUI compatibility.
If it does, I can start coding a graphical user interface, or GUI for you which will be much nicer than that little text box.
If not, then we work with the text box.
6-String Bass
July 24, 2008
Well, guess what I've been up to?
I added a new feature to the GMS program that converts a six-string guitar into a six-string bass guitar.
This is indeed very cool.
I was just jamming a little with it, sort of testing it out in my own crude way.
The bass sounds like a bass, what can I say?
It's just a simple pitch shifter that's shifting down to a ratio of approximately 30/82.
That's the frequencies of the low string on a 6-string bass and a 6-string.
Well, it turns out that through the magical mathematics of music, when you get that string right, all the other strings line up in place.
So you really do get a bass guitar sound out of it, down to the note.
What a simple, fun, easy effect to do.
Now if I could just get wah and tremelo figured out, I'd be golden, haha.
Guitar Handle
July 29, 2008
If you're following the project, you've noticed that things have slowed down a bit.
Well, we had a bit of a party in one of my chatrooms for two days and everyone got together at a meet, etc. was a blast.
I'm kind of in a waiting mode on the project.
The software is ready for adding the accelerometer interface to it, but the hardware is not ready.
Andy from Ultimarc is just receiving the prototype parts now, so he's got to solder it up and test it first.
I can't buy Wiimote parts just yet, so ixne on that for a while.
So what did I do?
Why, i drilled a handle into my guitar, of course!
I know, I know, it will alter the tone but that's worth it because I've thought up something fascinating.
You see, once we have the accelerometer creating effects variations in three dimensions and all that jazz, one style of playing that will emerge is that of dance.
Dance?
Yes, dance.
Imagine grabbing the guitar by a handle, strumming it with your other hand, then sweeping the guitar through an arc of muscle motion.
All the while the guitar is reacting, doing some "whoosh" or "whinddd" or something with the audio.
Neat, eh?
A person could develop a choreography routine with moves like Conan the Barbarian Sword Practice, haha.
Or advanced gymansts could run across the stage doing somersaults and wild stuff.
Of course, we still have the guitar wires to contend with, so we can have either a wired or a wireless sytem or both, and choose accordingly.
Anyway, just a little forward thought and that's what I've been up to, cutting a hole in my guitar.
Not much else going on, catch you later. - Les
Preparations
August 5, 2008
Yeah, preparations, that's what they are.
I have my software ready for testing, but I don't have any accelerometer hardware, so I'm kind of in a waiting pattern.
This seemed like a good time to practice with the guitar on the exercise moves that I had planned.
So I cleared out some space in the living room and started going through the motions without the guitar first.
And i was like stick-man!
My old bones wouldn't flex around the way they used to, and I found myself having enough trouble making the moves even without the guitar.
Obviously I'm in no shape for this.
So I did some light duty exercise and vowed to make it a daily activity until I am ready to comfortably heft around the guitar.
And it's a good thing I have no hardware yet cause I can't even make the moves yet.
In the mean time I studied some music.
I want to create a song using the guitar exercise routine, so I'm listening to some Funk and Heavy Metal songs, the foundations of the genres.
This will help me to express myself with those genres in some now-undetermined way.
I suppose I could get a WiiMote and start that part of the project up, but I'd prefer to wait a few more weeks for finances to really clear up better.
Sigh.
Patience is a virtue...
Progress
August 6, 2008
I sent Andy of Ultimarc a hello greeting with an update of my progress on the project, and he had good news.
He's currently waiting for a demo board from Analog Devices, and once he gets that he'll wire up a quick prototype, test it, and ship it right to me for testing and integration with software.
So a slight delay there but no problems so far and things are streamlined to get the concept into the DIY community as quickly as possible.
In the mean time I continue with my stretching and exercise routine - still not working up a sweat, just stretching old bones you know.
I played around with feedback today, putting the amp on distortion and cranking it up.
I then waved the guitar around, picking up feedback and sustaining it, stopping it, controlling it a little bit.
My thoughts are that a proper effects chain in such feedback could be useful as part of the guitar exercise concept.
For example, you can "pick up" some feedback by reaching the guitar into the amp, and then sustain it by pulling it out of the amp and retreating 10 feet back from it.
Finally, you can press on the strings to stop it.
Being in motion like that is facilitated by the exercise guitar concept, so it's possible to work those together.
One could swing the guitar in circles that are close to the amp as one idea.
Anyway, just some news to keep you posted.
Received Boards
August 9, 2008
Andy from Ultimarc has informed me via email that he has now received the development boards from Analog Devices and he will be able to ship me a working prototype within two weeks or sooner!
This is great news for the GMS project, and we wish Andy the best of speedy results in his efforts!
Prototype Ready
August 12, 2008
I'm very happy to report to you that the initial prototype is constructed and ready for testing!
Andy of Ultimarc will mail the prototype to me tomorrow and upon receipt I will run a test program on it to see how well it works with the software.
So great news, things are progressing rapidly!
Enclosures in the Works
August 15, 2008
I spoke with Wink and he seems to be interested in providing enclosures for us.
Naturally, in the DIY spirit of it all, you are free to create whatever enclosure you want for the device.
This is just for those who would like a more complete kit for assembly.
It will have three switches and will mount to the guitar in a convenient way.
To clarify, the accelerometer board will work just fine without any switches.
In that case you would just use the computer keyboard to set things up.
If you do want switches, there will be three configurable switches to do whatever you want to do.
Presumably this would be one switch for freeze and two switches for control, but you could set it up differently.
You might want all three switches for presets for example.
It is all up to how the software is created, so let me know what you want.
DHL Shipping Prototype and GMS15.ck
August 19, 2008
Just yesterday I received notice that the prototype is being shipped via DHL from the UK to Florida, and I'll be tracking it so I can test it as soon as it is ready.
Also I have just released revision 15 of the GMS software for you to download and enjoy.
It has more effects, a couple of bug fixes, and Betty voice feedback.
Operate the controls with the two shift keys, left to switch stages and right to change effects.
Prototype Working!
August 19, 2008
I have excellent news!
Today I received the prototype GMS sensor from DHL, as shown in the photo to the right.
They were real troopers, getting the shipment here in record time despite it being the day after a hurricane just hit us in Miami Florida, USA.
My hat's off to DHL for the delivery job.
Also my compliments go to Andy and his crew at Ultimarc for doing such a neat, clean, professional job on the wiring and soldering required.
As it turns out, they bought a different prototype than the one I suggested, so there was need to add a voltage regulator with supply cap, which was done quite nicely.
Some picky details are that the wiring harness is too long and the USB cable is too short, but hey - it's just a prototype!
We will optimize details like that for the final product.
Check out the second photo to the right to see how I mounted it to the guitar.
I just happened to have some double-sided foam tape from 3M (Scotch brand tape), which is super-sticky and super-strong.
Plus it has a thick piece of foam to absorb the surface irregularities of the thru-hole connectors and such.
So I cut two pieces to size, one for the U-HID Nano board and one for the sensor board, then stuck them in appropriate locations on the guitar.
I also used four pieces of electrical tape to mount the extra-long wiring harness on the guitar.
Although it would be easy to modify the length of the cable, I figured why mess with it and possibly ruin it?
Better to just mount it as-is for testing purposes.
So the initial prototype turns out to work perfectly without any enclosure at all, go figure.
Also there are three available switch connections which I am not using.
I prefer to operate the software from the keyboard for now and add switches / enclosure at a later date.
Oh, and I tried it out too!
It was so very satisfying to cue up reverb on the X-axis and pitch shifting on the Y-axis and just kind of move around while strumming the open strings to hear the sound vary.
I can't tell you how fortunate I feel that everything worked out so well.
Next there are some details to iron out regarding the design, but Andy assures me that he can spin a board fairly quickly to get the product out to you very soon.
Which of course is one of the reasons I selected Ultimarc and invited them to create the product - they are perfectly positioned to do this for us.
If I were to do it myself I would have to jump through a lot of hoops to re-learn software and get back into circuit board design which I used to do for a living.
Plus I have little desire to be a retailer - I like creating things and then seeing where they go.
Oh, and one more thing.
I'm modifying the software to subtract out the gravity vector and provide acceleration, velocity, and position as control options.
Also I added the space bar key to the GMS software so that we can set which sensor channel controls which effects stage.
More on that later.
At any rate, long story short the prototype works perfectly and I am one happy camper. Soon you will be too!
Details, Details
August 22, 2008
While it is true that the prototype is working just fine, there are some details to work out.
Specifically, the circuit is too noisy.
I added a little code that prints out some of the samples obtained from the prototype and set the guitar still in the guitar stand.
Since it wasn't moving, the printed values should be fairly constant but they were not, changing a little bit in the second decimal place.
This equated to an accuracy of only plus or minus five percent or so, which is clearly unacceptable.
In an attempt to fix this problem, I shortened the wires at Andy's suggestion which I agreed might help the problem, but it didn't.
Since I had the prototype off of the guitar for soldering of the wire harness, I decided this would be a good time to relocate it to a safer position.
You see, mounting the U-HID Nano board below the jack plate put the USB cable dangerously close to the guitar stand support rod, which was an accident waiting to happen.
So I put it above the jack plate and that works much better, as you can see in the photo to the right.
The coil of wire you see in the photo is the connection to the optional three buttons / switches and I left that untrimmed for future use.
Next Andy thought of the idea that we need larger capacitance on the output lines, which makes perfect sense to me.
Besides we need those caps to be larger to limit the bandwidth of the output signal and thus prevent any possible aliasing.
So that is where we stand right now on the hardware, still buried in some of the details of refining the design.
Software wise, I spoke about the software with some experienced guitarists on one of the guitar forums and I was told to remember to keep it simple.
Apparently I was making the software overly complex, which is true I really was.
So I'm currently working on simplifying the control interface of the software.
For example, I am eliminating velocity sensing as it was deemed that acceleration and position will be sufficient.
Also I am eliminating the X, Y, and Z options in favor of only theta and phi options because nobody really moves around orthogonally like a robot, but rather guitarists rotate around and such, so angles make more sense.
That's it for now, I'll keep you posted on details as they develop.
Noise Problem Solved
August 25, 2008
Well, the first and only problem is resolved.
The larger capacitors reduced the noise to a very tolerable level, so we are good to go with the design.
The only remaining factors are discussions between Andy and myself regarding whether the board should be separate or integrated, cable length, mouning hardware, and the like.
So once we've got that all ironed out we will be ready to make some circuit boards!
They'll be available soon!
Board Design Complete
September 3, 2008
More good news!
The circuit boards have been designed and sent to the manufacturer, so we should have the initial batch within a week or so!
It's been a journey of several months of planning plus maybe 6 weeks of actual work to get the product out to you, and I've enjoyed it immensely.
Now The software is in pretty good shape, but I need to finalize some sensitivity issues and test with the final product before releasing the next version of it.
There has also been some talk of a USB to MIDI converter that would allow you to skip my software completely and use your favorite effects software with this device.
As always, I'll keep you posted!
Software Professional On Board
September 4, 2008
I am very pleased to announce that we have a new software tester / developer on board the project!
As a professional driver developer, he is uniquely skilled in exactly the area of expertise that we need.
He will be looking at existing USB to MIDI drivers especially with regard to interfacing the GMS board to existing VST plug-ins.
This will free you up from needing to use my cheezy amateur software and open up the world of VST effects software, some of which you may already be using.
As a driver developer, he can also code up custom drivers if required, which would enable features like buttons, switches, sensing the gravity vector, and things of that nature.
So now it seems like this little project may be growing beyond my expertise and my role as a hobbyist technology developer may be shifting to more of a writer and web focal point for the guitarist community.
I couldn't be happier!
New Software
September 12, 2008
I have upgraded the GMS to rev 30 now, with a new more complete Betty voice file.
This is all fine and dandy for experimenting and trying all possible combinations of effects and such, but in practice I found it rather confusing.
For one thing the voice feedback, though cool at first, gets annoying.
For another all those stages with all those menu options can create predicaments, not to mention an extremeness mouse adjustment for each stage change.
Finally and most importantly, the CPU burden is a bit much for my 4-year old slowish computer, so it will likely be too much for your machine as well.
So what are we going to do?
Easy: write some simplified, streamlined software that does just what we need and nothing else.
So I've done this for the wah effect and it turned out pretty good.
The resulting program is Wah5.ck, which is available on the download page.
Look for more streamlined software as time permits me to write it.
More New Software
September 13, 2008
I've been at work lately and produced two more of these simplified programs that don't have any complicated control interface.
They both work by sensing the guitar's body vibration, which the accelerometer can do, and then altering the sound with that.
The first is Pitch1.ck, which makes sort of a "Galaga Bee" type of sound that is kind of interesting, and the second is Vibes1.ck, which is really cool.
Vibes1.ck has a rich, sustained quality to the sound that I really like.
So give these a try once you have a motion sensor on your guitar, you'll like them.
Finally Boards!
October 25, 2008
I must apologize for the lack of action on our part, but simply stated the boards were "in production".
Andy chose to go with his Indian manufacturing company which is low cost but slow, so we had to wait - but the good news is the boards will be reasonably priced because of this.
On Monday (in two days), Andy will receive the first 5 sample circuit boards, so things should start happening soon.
Chip Delay
November 7, 2008
We have run into a minor snag, apparently Analog Devices has put the ADXL330 chip on 8-weeks worldwide allocation delay.
Unfortunately this means that the accelerometer boards will not be available for an additional time of up to 8 weeks.
Nothing can be done about this as it is simply due to the high demand of the chips.
I'll keep you posted on any changes.